Introduction: The framework for Cash & Voucher interventions

The appropriateness of cash and voucher-based interventions has become an increasingly discussed topic in the field of humanitarian assistance. Practitioners conventionally support the direct distribution of food, non-food, and agricultural commodities to meet the immediate emergency and recovery needs of households and communities impacted by natural or man-made disasters. The direct distribution methodology aims to assist beneficiaries with the necessary items to survive and recover from a crisis.

Cash and voucher approaches, through which needy households can gain access to household items or food stuffs to improve household welfare, have long been popular in Europe and the United States. Examples from the United States include food stamps and rent subsidies for the poor or dollar-denominated credit cards as were issued to victims of Hurricane Katrina. In the developing world, cash and vouchers have sometimes been used to promote household access to seed, non-food items, food, and labor.

The direct distribution approach often cannot accommodate the unique needs of a recipient household. This limitation of direct distributions often leads recipients to sell or trade certain items in order to acquire items of greater value to the household. When trade or sale is not feasible, beneficiaries of direct distributions may keep and use the articles received but not derive the same utility that trade would engender. When items are sold, the transaction tends to occur under unfavorable terms of trade for the seller, with the resulting sale price below the local market price and often below even the procurement price for the item.

The common practice of beneficiaries trading or selling direct distributed assistance such as food or non-food items may give the impression that assistance is unnecessary, when in fact it is the ‘type of assistance’ that is not appropriate to the situation.

Direct distributions, particularly where procured commodities are from countries and regions outside of the area being assisted, can create market distortions and negatively impact recovering economies by discouraging local supply.

The principle argument for voucher and cash-based interventions is that beneficiaries can access the goods and services, if available, responding to their individual and household priorities while also supporting economic recovery by making purchases locally and in the process stimulating and supporting local markets.

Voucher and cash interventions present their own potential negative externalities.1 Cash interventions could potentially create local inflation, to the disadvantage of non-beneficiaries to the voucher or cash schemes. The use of cash can also present security risks for beneficiaries and staff. Beneficiaries might not use the money in optimal ways to maximize a household’s welfare. However, after an extensive review of cash and voucher activities, Oxfam found that fears about local inflation, purchasing of non-essential commodities, and security risks are generally not realized in practice or that the risks are no greater than those inherent with in-kind distributions.2

A final concern about cash and vouchers involves gender and intra-household control of resources. As heads of households usually control productive resources such as cash,3 and women are more likely to have control over food resources,4 cash-based activities could potentially disadvantage women. This points to the importance of gender biasing interventions to ensure that women are empowered to make decisions relative to household allocation of resources. Gender biasing is based on the hypothesis that men tend to make decisions without spousal consent within a household and that when women are empowered with choice, male decision making is less oriented than their female counterparts towards maximizing household welfare.5

Evidence from voucher-based interventions in Central Africa indicates a tendency for men to increasingly take interest and control of the use of vouchers. During the course of three successive voucher interventions in a single community over a 13 month period, where there were no changes in the selection criteria of households chosen to receive vouchers, the percentage of vouchers that were utilized by men at the point of trade increased. During the first voucher intervention in the community, voucher users were equally split between women and men. By the third voucher intervention 13 months later men accounted for 68% of voucher users.6 Somewhat in contrast, Oxfam found that designating women as recipients of cash or vouchers actually increased their decision-making authority within the household by increasing their share of income.7

Agency, Country, and Project Context of the intervention

Catholic Relief Services (CRS), the overseas relief and development agency of the United States Catholic Community, has a 60 year history of responding to emergencies in developing countries. While historically CRS’s emergency interventions have tended to be direct distributions (food, seed, and non-food items), over the past ten years the agency has developed an institutional capacity with using vouchers. CRS’s voucher work is primarily in seed, Seed Vouchers & Fairs,8 which have been employed in over 25 countries in Africa, Asia, and Latin America. CRS voucher experience includes the use of voucher to access non-food items in Afghanistan, Iraq, and Indonesia as well as the use of labor vouchers in Uganda and Ethiopia.

The CRS program in the Democratic Republic of the Congo (DRC) piloted the use of vouchers to promote access to seed in DRC starting in 2004 through training and support to a number of local partners. The experience with Seed Vouchers & Fairs prompted CRS and local partners to expand the use of this approach. By early 2005, CRS Congo had submitted emergency response proposals to the Gates Foundation, UNICEF (United Nations Children’s Fund), and OCHA (United Nations Office for Coordination of Humanitarian Affairs) to provide humanitarian assistance in Eastern DRC with a focus on a complex emergency in Katanga Province. CRS Congo’s proposals to both donors included programming requests to fund emergency seed and non-food item support to vulnerable households and to use, where appropriate, voucher approaches.9

CRS Congo, in partnership with local partner Caritas Kindu, implemented a post-emergency livelihood project in Maniema Province, Eastern Democratic Republic of Congo, from January 2004 until April 2006. The project addressed seed, infrastructure, and non-food item needs through Seed Voucher & Fairs, cash for work to support road rehabilitation, and non-food item (NFI) distributions of pre-packaged UNICEF household kits.

The UNICEF kits contain two metal cooking pots, four hard plastic plates, four cups, eating utensils, one plastic sheet, two blankets, a hard sided 20 liter jerry can, and 500 grams of soap. CRS’s program in Kindu supplemented these kits with agricultural tools and used clothing. While monitoring and evaluation has shown that beneficiaries consider these items useful and are still using them three months after their distribution, neither M&E nor initial needs assessment evaluations can determine the value that households place on certain items or the extent to which these items would have been procured by recipients if they had been given choice and access to a wide variety of items.

Maniema’s population experienced massive displacement during the two wars that took place in Congo between 1998 and 2003. Sexual violence and indiscriminate killing of civilians was rampant. Household looting and pillaging was widespread. Since 2003, relative stability has prevailed in the province and families have returned to their farms and villages. Most returnee families had the additional challenge of re-establishing their livelihoods in the face of destroyed and looted houses.

CRS distributed NFI kits more than two years after fighting had ceased because evaluations indicated that a high need for non-food items still existed at the end of the first phase of programming in 2005. CRS therefore continued with NFI kits activities in 2005 and 2006, targeting more isolated areas not reached by the initial activities.

Objectives of the Study

In order evaluate the appropriateness of its NFI activities and illuminate the advantages and potential pitfalls of using cash and voucher interventions to meet the non-food needs of Congolese households recovering from the effects of war and displacement, CRS Congo used private funds to carry out a pilot ‘cash for household non-food needs’ program in Maniema Province, DRC in April 2006. The pilot involved 40 households and took place over a four week period from the time that households were selected until the day that they made their purchases. Households were provided clear guidelines as to what could and could not be purchased with the cash, and they only had one market day in order to purchase the items. Cash recipients were selected through a lottery of a sub-group that self-selected for participation in the study. The time limitation, purchasing rules, and selection process were used to facilitate research and provide a direct comparison with the on-going NFI activities while also not disadvantaging beneficiaries, and would be modified in a true cash and voucher program. Cash recipients were provided with the local currency equivalent of $62.56 and were monitored closely to document and understand their purchase decisions. From among the areas already identified for the NFI distributions, the villages selected for participation in the pilot were chosen for their relative proximity to a major market.

The pilot objectives included:

  1. To understand the actual household non-food needs of households selected for the UNICEF NFI kit program implemented by CRS and Caritas Kindu.
  2. To assess the extent to which the UNICEF NFI kit was meeting household needs as typically a household’s ‘expressed need’ differs from their ‘demonstrated need,’ the latter evidenced by what the household actually procures to meet its welfare needs when given the choice.
  3. To understand intra-household decision making, notably that of households that were targeted as being war-impacted and in need of non-food assistance.

While issues regarding inflation and security are beyond the scope of the pilot and the limited scope presents biases which may limit the applicability of findings to other contexts, the pilot sheds light on intra-household decision-making dynamics, the appropriateness of the NFI kit intervention, and the items in demand by families in this post-conflict setting.


Selection of participants

Participants were selected in February 2006, among households from three villages 24-40 km from the provincial capital of Kindu. The villages were chosen in November 2005 for an NFI intervention for April 2006. Inhabitants in these villages had been displaced several times during the conflicts in DRC and had returned to their villages in 2003 and 2004. Households from three villages were given the option of entering their name in a lottery to participate in the study instead of receiving an NFI kit. The field staff explained the rules of the study and emphasized that participation was optional. Staff also explained that households would not be given a choice as to whether their household would be represented by the husband or wife. Forty households participated in the study. Men, who in Congo are always considered the head of household in two-parent families, were to represent twenty of households.

The selection process introduced three biases. First, areas closer to major markets are better off than those that are not because this facilitates evacuation of agricultural goods. Second, the self selection process results in entrepreneurial participants who already find cash advantageous. Third, it forces the participation of women who otherwise would be represented by their husbands.

The staff estimates that 60% of those present volunteered to participate in the study, the date of which had been set for three weeks from the date of participant selection. The primary reason for the reluctance among beneficiaries to volunteer was their concern that the money that they would receive would be less than the kit’s true value. The names were chosen at random in front of the villagers to ensure that the selection process was transparent. The date of the market trip was set for just prior to that of the NFI distribution in order to not discourage participation in the pilot.

One of the men selected had two wives, and the rest were from monogamous families. For the households represented by women, twelve were from two-parent monogamous families and one was a widow with children. Six of the women who participated were first and second wives with common husbands, and one of the women was a second wife (the first wife will receive an NFI kit).10

Study methodology

Each participant was given the local currency equivalent of the kit’s value as priced in Kindu,11 which has numerous stores and market stalls where household goods are sold. Because the same quality of plastic sheeting is normally not available in Kindu, the price of $15 was decided upon for this item.12 The plastic sheeting, along with all other kit items, was available “for sale” from CRS, in case items of identical quality were not readily available in the market. Participants therefore had the option of trading in all of their cash and receiving a UNICEF NFI kit.


The participants were taken from their villages by truck to Kindu. The participants were notified a week prior to the market trip to confirm the date. The day that the truck arrived to take the participants to Kindu, four cases occurred where men presented themselves for households that had been designated to be represented by women. The men were explained that they had the choice of sending their wives or opting for the kit instead of the cash. In all cases the men chose to send their wives.

The study participants had three hours to spend their money on non-food items in Kindu, following these criteria:

  • Only non-food items could be purchased (such as clothing, tools, kitchen items, bicycles, and radios). Food, livestock, and seed could not be purchased.
  • Caustic and flammable items could not be purchased (including fuel).
  • All cash must be spent.
  • Participants could combine their money with each other.
    Participants could add their personal money.

The criteria were read aloud to participants the day of the study, though they had already been informed during the participant selection that their purchases would be limited to non-food items. While combining money between participants was allowed, this was not stated to beneficiaries, as they had not been told in advance and the introduction of this element at the last minute might have effected their pre-planned decisions. Staff accompanied the participants (2-3 participants per staff member) in order to answer questions on allowable items and ensure that participants followed the parameters of the study. They also recorded the items bought and their prices. Upon completion of their purchases, participants were interviewed individually regarding why they selected their items, the process by which they made their decisions, and the overall experience of using cash instead of receiving goods.

While the study involves the distribution of cash to participants, it is essentially a study on vouchers, because the purchasing is limited to certain items (no food, livestock, seed, or caustic/flammable commodities). A true cash study would not place restrictions on the use of the money or the time frame in which it had to be spent. However, in order also to evaluate the appropriateness of its NFI activities in terms of the pre-packaged kit in comparison with a voucher system, the decision was taken to restrict the purchasing to non-food items. This narrows the focus of the study to the non-food commodities specifically in demand, as opposed to what overall commodities were in demand by participating households.

Results: What was acquired, gender variances, intra-household consultation

The purchasing decisions of participants testify to the trade-offs made to meet household priorities and to the appropriateness of the pre-packaged NFI kit. Several trends were apparent: the prevalence of clothing, pagnes, and mattresses, the relative absence of NFI kit items, and the selection of goods for all members of the household.


Most beneficiaries (95%) bought at least one article of clothing or shoes. Three-quarters of beneficiaries purchased a pagne (a six-meter length of material used primarily by women for clothing) with 83% of those buying pagnes opting for the high quality brand. Sixty percent purchased mattresses, and half bought one or more kitchen articles.

A typical basket of goods acquired was:

  • 1 mattress OR Bicycle parts – $29
  • 1 quality pagne – $10
  • 3 articles clothing (new and used, including at least one children’s) – $8.33
  • Suitcase/bag/trunk OR radio $9.65
  • Soap – $.33
  • Kitchen items – $1.50
  • Plastic sandals – $3.00
  • Batteries – $.25
  • Unaccounted for – $.50

Certain items invariably cost more than others. Mattresses and bicycles parts were the most expensive items. Mattresses ($29) required participants to spend nearly half of their money for this single purchase. The average amount spent on bicycle parts was $33.13 When the total amount of money spent by all participants is considered, mattress and clothing/shoes were the top two investments. Clothing and shoes vary greatly in price depending on their quality and whether they are new or used, and those who bought clothing spent an average of $13.62. Pagnes, which cost $10 for high quality and $6.50 for lower quality pieces, and bicycle parts were the third and forth largest expenditures. Mattresses, clothing/shoes, pagnes, and bicycle parts and accounted for 79% of money spent during the study.


Overall, participants spent only 8% of their money on kit items. The most commonly purchased kit item was used clothing, followed by plates. While both of these items were available to be purchased from CRS, beneficiaries opted to find them in the market in order to select the size, quality and quantity of their choice. The only exception was a participant who bought a blanket from CRS. She stated that she had not planned on buying a blanket but she made the purchase because it was high quality.

One quarter of the participants wanted to purchase an unauthorized product – namely salt, sugar or fuel – or keep some money to pay for school fees. When purchase receipts were tallied, 0.4% of the money was unaccounted for, presumably pocketed by participants.

Concerning purchases that might be considered non-essential or even frivolous by humanitarian actors, 0.3% of total spending went to beauty products, 0.4% to cassettes, 0.8% to games, 0.1% to mirrors, 0.1% to watches, and 0.02% to shoe polish.


A key question for the study was whether the purchases would vary between men and women representing households. In fact, the differences were negligible. For both men and women, the items accounting for the four largest expenditures were mattresses, pagnes, bicycle parts, and clothing/shoes. However, for women, clothing and shoes were the most significant expenditure (25% of total spending), while for men it was their fourth largest (13%).


Study participants were asked to designate the recipient of each good. Bicycles, radios, cassettes, batteries, men’s clothing, shoes and bags, and tools went to the male of the house; women’s clothing, shoes, bags, and pagnes went to the woman of the house, and children’s clothing and shoes were the primary products bought for children. Goods going to the household at large were mattresses, kitchen articles, trunks, locks, towels, mosquito nets, and umbrellas. Both men and women spent half of their money on products for the household and children, and they spent the remaining half on products destined to the adults of the household. Women spent more on their children (14% as opposed to 4% spent by men) and men spent more on general household goods (49%) than women (36%). Women spent a quarter of their money on products for themselves and one quarter on products for their husbands. Men spent 30% of their money on products for themselves and 17% on products for their wives.

The question as to which sex spent more of their money geared towards to overall household welfare is inconclusive. Women did spend more money on children’s clothing and shoes. Seventy percent of the clothing expenditures made by men were on men’s clothing, while women’s spending was geared towards children’s clothing (55%) and men’s clothes (25%). However, half of the money that women spent on children’s clothing, and therefore a significant part of their clothing expenses, was on “luxury” clothing. It could be argued that purchasing more expensive dresses and clothing sets is a frivolous expense when less expensive used clothing is available. Men and women spent almost equally on bicycles (used for transporting agricultural products) and tools, which were the only productive assets purchased. Men spent more on mattresses, which benefit the wife and children as well. Overall, the study does not support the claim that men would be more inclined to purchase goods that do not necessarily benefit the household, but it does show that purchases by men and women may echo the dominant gender roles in the household.


The similarities in purchases made by men and women reflect the planning that was done between the spouses in the household before the study. All married participants stated that that had consulted their spouses regarding the purchases, and one-third also consulted their children. Interviews with participants revealed that the planning process within the household involved discussions on needs and priorities, where the wife was very much involved in the decision-making process but that the husband typically had the final decision on what items would be purchased. Participants described making lists of items that they were missing and then deciding with their spouses what they would buy. More than a third of participants showed up for the study with written lists.

The husbands of just more than half of the women participants found their own transport to Kindu in order to assist their wives with purchases. Four wives of male participants also showed up in Kindu the day of the study. One helped her husband, who was illiterate, and the others came in order to select their pagnes. The fact that four husbands tried to go in place of their wives and that eleven showed up the day of the study can be interpreted as either rent-seeking behavior, mistrust, or concern wives may be swindled or have trouble managing the large amount of cash. Staff present at the study felt that the presence of the husbands appeared to be more a case of them helping their spouses than supervising them. The husbands usually scouted out for goods and wanted to make sure that the women were not taken advantage of by vendors.

About one-quarter of the participants wanted to buy something that his or her spouse was against. One man, who brought a bike frame, described that his wife proposed he buy a pair of new pants for himself instead of a quality pagne for her, but that he refused and bought the pagne. In another case a woman wanted to buy a mosquito net and a gas lamp, but her husband dismissed these ideas.

The decision-making process came across as a positive experience that enabled a problem analysis that might not otherwise take place and an open discussion about how to best meet the needs of the household. Several female participants expressed that the experience made them feel more involved in household decisions. In describing his discussion with his wife, one man stated:

At first my wife wanted the kit, but then I showed her that we already had a few plates and pots, and that we really needed a bicycle because it is a source of production. I said that we could buy clothes with the money that we will make. She accepted the idea of the money and the bicycle.

Another man, describing why he preferred the study method to receiving the NFI kit, noted that the planning process was an advantage:

In consulting with my wife before coming, this allowed us to know the list of our needs that we were unaware of.

A female participant described the planning done with her husband that took into account the political climate of the Congo:

After doing an inventory of our goods, it was apparent that we lacked a good place to sleep…and also a way for my husband to listen to get news, especially during this period of elections, so we preferred to have a radio…

For the pairs of first and second wives that participated in the study received, each received separate amounts of money. However, the planning process was one where they worked together with their husband on how they would spend their money. One pair combined their money to purchase a bicycle, stating that their numerous needs within the households would be better served from the income that their husband would eventually make with the bicycle. In one case, the second wife from a polygamous family participated on her own (the other wife will receive an NFI kit). She described the planning process, during which she, the “rival” wife and the husband devised a strategy to make best use of the kit:

We all got together, my husband, my rival and me. My husband decided that since my rival would receive the kit, that I should take advantage of this to buy items that were not included, except for plates. We did all this after having discussed priorities in our house. The clothes are for my children and my rival’s children, and my husband said that I could buy a pagne because my rival already had one. Therefore, my purchases are the decision of my husband, my rival and me.

While evident that the husband had the last word on purchases, every female participant preferred the cash over the kit, indicating that exclusion from the decision-making process was not a significant problem and that they ultimately agreed with the purchasing decisions, at least in comparison to the contents of the kit.

It appears that the large amount of money given to participants may have influenced them to make larger purchases. One male participant said that, “I proposed to my wife that we purchase clothing for the children. She preferred that we buy higher value objects now and that the children’s clothing we could purchase later.” Considering that mattresses and bicycle parts cost $29-$32 and upwards – an amount that could take months for a household save – the opportunity to purchase a big-ticket item was probably seen as too good to pass up. The cash lump-sum enabled beneficiaries to purchase certain items that might otherwise be unattainable.

The participants spent a total of $2,497 among dozens of local vendors and shops. They did not report any problems of vendors charging them higher than normal prices.

Key findings

The post-emergency context of Maniema provides an interesting setting for the study, which was conducted nearly three years after fighting had ceased. In an acute emergency, practitioners may assume that the gravity of needs will lead to purchases that will necessarily be directed towards promoting household welfare, as OXFAM assumed in its cash distribution program in Somalia in 2003-2004.14 The risk for the purchase of non-essential commodities, more generally accessed through income-generating activities or micro-finance programs than emergency programs, would logically be higher in the context of Maniema than that of an acute emergency. The results of the Kindu cash for non-food items pilot show that NFI kit items were not in strong demand by study participants. When queried, participants indicated that they already had some of the non-food items that were available in the NFI kit, such as pots, and that they no longer had use for some NFI items such as plastic sheeting, which is typically used to repair a leaky roof or to serve as the basis for a temporary shelter.

Due to the rules limiting participant purchases to non-food items and the structured and heavily monitored environment under which the pilot was executed, it is difficult to ascertain if and how the participants spending patterns would have differed under less controlled environment. It is presumed that the controlled nature of the study discouraged the purchase of items not deemed critical to household welfare.

Even with rules confining purchases, all forty participants stated that they overwhelmingly preferred method of receiving cash and making purchases in the Kindu market rather than receiving the NFI kit, because it allowed them the flexibility to respond to their priorities, though this finding is unsurprising as the participants themselves opted to receive cash instead of the kit. They preferred being able to purchase items in the market of the quality and quantity of their choice, even when they purchased kit items. Providing cash rather than the NFI kit encouraged a needs identification process within the household that may not have otherwise taken place.

The study highlights the challenges of imposing or promoting non-food items deemed by aid organizations as being most beneficial to household welfare, such as soap and mosquito nets, which can improve household hygiene and health. For instance, it could be questioned whether the significant amounts of money spend on quality pagnes and new clothes for children are “proper” investments, when mosquito nets could reduce their risk of contracting malaria. Ultimately, cash and vouchers empower households to make the choices they feel most beneficial to their physical, social and psychological welfare. The designers of the study deliberately did not “ban” more frivolous items like lipstick, film, and shoe polish for two reasons. First, they wanted households to have the option of purchasing the items and to monitor the extent to which they exercised this option (mirrors, beauty products, shoe-polish, cassettes, and games accounted for less than 2% of spending). Second, limits would prevent households from maximizing utility. This issue also taps into the challenges of how donors and practitioners evaluate the appropriateness of purchases of beneficiaries “empowered” through cash and vouchers: certain items, such as alcohol and arms, are clearly detrimental to household welfare. Other non-essential items like games and beauty products create a grey area of social and psychological benefits.

The results of the CRS Kindu cash for non-food item study show that the pre-packaged NFI kits were not appropriate for the area at the time of the pilot and as participants spent less than 8% of cash to acquire non-food items that were in the NFI kits.

It is important to note that the villages selected for inclusion in this study were chosen for their access to the market. As a result, households here had far greater market access, if not buying power, to acquire many of the household items in the NFI kit as compared to more isolated villages. Villages where the study took place benefited from a Seed Voucher & Fairs in 2004, and some of the study participants gained access to seeds and tools through vouchers. The length of time since the acute crisis, the proximity of the participating village to major markets, and the fact that study participants had already benefited from other livelihood interventions all contributed to a diminished need for the basic items found in the NFI kit as evidenced by the study.

One challenge with conventional NFI interventions is the logistics involved with transporting large numbers of bulky kits. Transporting conventional NFI kits to areas more than 100 kilometers from a passable road proved unrealistic financially and logistically. Needs assessment evaluations conducted in the Maniema NFI distribution activities revealed that the more isolated areas in the province had greater needs than less isolated areas, but many areas could not be served with NFI kits because of the lack of roads.

Villages within forty or fifty kilometers of a major market center like Kindu have more opportunities to access markets, acquiring cash by selling their agricultural products, than villages that are cut-off because of long distances and impassable roads. The results of the study likely would have been different if the participants had been from villages that were a few days journey from the nearest market. The study was only feasible because a robust market with diverse non-food items was accessible to the target area. The bias to assist accessible areas was therefore present in both the NFI kit distribution activities and the cash for non-food items pilot.


  1. Doing a cash or voucher intervention for NFI requires a viable and ready market for NFI which is typically lacking in the rural DRC context. The challenge of ensuring supply becomes even more apparent when one considers that complex emergencies like the one in the Congo may result in the destruction of a functioning market system. The absence of items may reflect a lack of demand or the failure of the market to supply items in demand. For a cash or voucher intervention to be successful, the program may have to intervene directly in the market by providing loans to vendors, subsidizing transport, or by acting as a seller. If vendors do not have sufficient stocks, choices available to beneficiaries will decrease, which would reduce the utility of cash or vouchers based approaches.
  2. Switching from direct distributions to cash and voucher interventions provides choice and empowerment to recipients, enables a system of delivery that promotes recourse and accountability, reduces project transport costs, and supports and stimulates local entrepreneurs and the local economy.
  3. Cash and voucher interventions put the beneficiary at the center of the process, empowering beneficiaries to choose the items they determine to be most necessary for their predicament. This choice promotes needs identification and encourages dialogue within the household concerning priorities that might not otherwise occur.
  4. Cash and voucher interventions must assess and incorporate gender dynamics in planning, execution, monitoring and evaluation, as control of cash household resources typical falls to the male of the household and purchases may reflect dominant gender roles in the household. This is particularly crucial because vouchers may represent assets which, if given through a direct distribution, would be controlled by women (i.e. food, cooking pots). Interventions should encourage the full participation of women in the household needs identification and purchasing processes, while recognizing that husbands will still likely control the cash or voucher, no matter who physically receives it. If monitoring reveals that control of vouchers by men within the household results in women and/or children not accessing items critical to the household, interventions can provide vouchers only valid for specific items, but this comes at the cost of limiting choice.
  5. Cash and voucher interventions also build social capital. This is both between sellers and buyers, which in many cases can be host and displaced populations where there is inherent tension, as well as between buyers. Vouchers can also allow households to pool their vouchers and procure items that might otherwise be unattainable to an individual household. Social capital tends to be low during times of crisis and efforts that strengthen social capital support the building of civil society, which is particularly important in post conflict environments.
  6. There are significant challenges to executing and funding cash or voucher-based NFI programs. Market and product access is a key limiting factor operationally. Donors that fund NFI programs are generally leery of the use of cash or vouchers as most NFI programs function in areas marked by conflict or natural disaster where markets are not functioning.
  7. For cash and voucher NFI interventions to be successful in empowering beneficiaries to meet their household needs, practitioners must find a way to bring people close to the market or to bring the market close to the people. Options here may include subsidizing trader or buyer’s transport costs, expanding the temporal validity of vouchers, or promoting availability of items deemed appropriate through up front contracts to suppliers. The study opted for the first option by providing transport to participants, though in a real project participants could have been asked to get there on their own. While cash and vouchers are not a universal solution to the challenges of traditional direct distributions methodologies, practitioners need to critically look at these approaches in addressing the needs of vulnerable households in both emergency and post-emergency contexts.


The Kindu pilot study was made possible through the support of the CRS Congo program. Special thanks to Dorothy Madison-Seck, CRS Regional Director who encouraged the study and to Ryan Russel, Head of Emergency Programs for CRS in DRC, who provided valuable insights into Kindu and challenges to conducting voucher based programs. This pilot would not have been possible without the leadership, commitment, and guidance of Caritas Kindu, CRS’s operating partner in Maniema Province, whose staff carried out the study. A note of thank you to UNICEF whose program support would not have made the comparative nature of this study possible. Lastly, thanks to the people of Maniema Province whose perseverance to the horrors of war inspired this study. The bar of service to the most needy must never be lowered by context.

About the Authors

The authors are both MALD graduates of The Fletcher School of Law & Diplomacy. Sarah Bailey has worked in the humanitarian field for three years in Niger, DRC, and Burundi, focusing on rapid emergency response in complex emergencies and livelihood recovery in post conflict contexts. She served as Emergency Program Manager for CRS in Maniema and Katanga, DRC. Stephen Walsh has worked in the humanitarian field since 1999 with CRS. He worked in complex humanitarian emergencies in Sudan, Burundi, and DRC gaining extensive experience in seed, food & non-food emergency programs and voucher backed approaches. He served as Head of Programming for CRS Congo during the time of the study.

  1. For a more complete discussion of possible advantages and disadvantages of cash and voucher programming in emergencies, see Harvey, Paul. 2005. Cash and vouchers in emergencies. HPG Discussion Paper. London: Humanitarian Practice Group; Harvey, Paul. 2007. Cash-based responses in emergencies. London: Humanitarian Practice Group; Peppiatt, David, John Mitchell, and Penny Holzmann. 2001. Cash transfers in emergencies: evaluating benefits and assessing risks. London: Humanitarian Practice Network; and Creti, Pantaleo and Jaspars, Susanne (eds). 2006. Cash Transfer Programming in Emergencies. Oxford: Oxfam GB. [return]
  2. Creti, Pantaleo and Jaspars, Susanne (eds). 2006. Cash Transfer Programming in Emergencies. Oxford: Oxfam GB. [return]
  3. Quisumbing, A.and McClafferty, B. 2006. Food Security in Practice No. 2 – Using Gender in Development. International Food Policy Research Institute (IFPRI). [return]
  4. Walsh sites studies by Katona-Apte (1986), Bryson et. al (1992) in Walsh, M. 1998. Women in Food Aid Interventions: Impacts and Issues. Institute of Development Studies. [return]
  5. Examples of studies showing that women tend to spend more money promoting household welfare and food security than their male counterparts include Kammi K Schmeer (2005), John Hoddinott and Lawrence Haddad (1995), M. Garcia (1990), and Duncan Thomas (1990). [return]
  6. Walsh, S., Ngendahayo, M. and Droeven, C. 2003. Placing the farming family at the center of the process through seed demand is met: CRS/Burundi’s experience with seed vouchers and fairs in Kirundo Province. [return]
  7. Creti , Pantaleo and Jaspars, Susanne (eds). 2006. Cash Transfer Programming in Emergencies. Oxford: Oxfam GB. [return]
  8. Remington, T., Maroko, J., Walsh, S., Omanga, P., and Charles, E. 2002. Getting off the seed-and-tools treadmill with CRS seed vouchers and fairs. Disasters 26(4):316–328. Available on-line at (accessed September 2004). [return]
  9. Both programs were successfully funded with the Gates project having more of an explicit voucher focus and the UNICEF project more oriented to supporting effective needs assessments and coordinating response. [return]
  10. The CRS project treats first, second and third wives and their offspring as separate households, since they live in separate houses and do not always share resources. [return]
  11. Field staff priced the items at three different merchants. The average of these prices was used. [return]
  12. The plastic sheeting distributed by UNICEF is of a much higher quality than plastic sheeting available locally. When it is available in the Kindu market, it costs $28, which is more than double the cost of this item in Goma. The study designers reasoned that in a project setting they would fix a price with vendors or act as a vendor, with $15 being a price that would cover purchasing the item and transporting it from Goma. [return]
  13. Bicycles can be bought unassembled, and the majority of those made bike purchases opted for only a frame and other key pieces, rather than the entire bicycle. [return]
  14. Degan Ali, Fanta Toure and Tilleke Kiewied (March 2005). Cash Relief in a Contested Area: Lessons from Somalia. Humanitarian Practice Network Paper #50. Overseas Development Institute: London. [return]

3 Responses to The Use of Cash in Emergency and Post-Emergency Non-Food Item Programs – A Case Study from the Democratic Republic of Congo

  1. Thank you for bringing attention to the many forms of need, it is a concern that we share.

  2. Colonel Warner Anderson says:

    I was a recipient of commodity foods and “food stamps” for 7 years while running free clinics, and during graduate and medical school. I find I now pay more in taxes in about 6 months than I received during all those years. Ideally, this is how a voucher program should work. Not all investments pay off, but it may only take relatively few pay-offs to reimburse and refresh the system.

    One wonders, noting the economic latitude created by polygamous households, if the next step is to increase purchase power and perhaps head off inevitable inflation by forming bargaining groupps to negotiate prices, or a “commune”-type grassroots organization to bring goods to the purchaser, rather than vice versa.

  3. Olivier Eyenga says:

    Very interresting study